"Alarming" increase in MRSA infections in children

I've been waiting to post on this paper from last week because there was something odd about the math in it. It's an important and troubling finding, though, so even though I haven't heard back from the authors, I'm going to describe it for you and explain where I think the error is. (NB, When corrected the error makes their findings more significant, not less.)

So: Authors from Emory University in Atlanta reviewed a national database of head and neck infections in children that were reported by 300 hospitals between 2001 and 2006. In those years, there were 21,009 S. aureus infections: ear infections, sinus infections and so on. But, in 2001, the proportion of S. aureus that was MRSA was 11.8%. In 2006, it was 28.1%. That's more than a doubling — it's a 138% increase. (Here's the error: The authors say it was a "16.3% increase". I believe they mean "16.3 percentage points," which would be correct but significantly undersells their finding.)

Because 60% of the infections were outpatient, the authors make the reasonable assumption that these are community-associated MRSA (CA-MRSA). CA-MRSA is generally resistant to fewer drug than the hospital (HA-MRSA) variety. Unfortunately, they also find that this community strain is gaining additional resistance factors: More than 48% of all MRSA stains in this study were resistant to clindamycin, which is not a traditional resistance for CA-MRSA but recently has been reported more and more. Also, head and neck infections are not traditional presentations for MRSA, suggesting the bug is expanding its range.

Their comment:

...[T]he results of this study depict an alarming increase in MRSA in the United States. There is an increasing trend of clindamycin resistance among MRSA isolates. Expeditious culture of suspected head and neck infections leading to more appropriate antimicrobial drug selection is highly recommended to avoid further resistant patterns.

The cite is: Iman Naseri; Robert C. Jerris; Steven E. Sobol. Nationwide Trends in Pediatric Staphylococcus aureus Head and Neck Infections. Arch Otolaryngol Head Neck Surg. 2009;135(1):14-16.